[R] Deleting observations - can't see the data after that

Schwab,Wilhelm K bschwab at anest.ufl.edu
Fri Oct 8 16:20:46 CEST 2010


Josh,

How about the direct approach: grep or otherwise search the source?  If you can guess the form of a printf()-like statement, then it might be easy to find.  More likely, one would have to overlay results of a few different searches.  Maybe someone who is actively working on the code will see this and stumble on it??  The idea is simply that one might be able to work backwards from the behavior (which we have both observed) to possible causes and give a little thought to possible absorbing states that maybe should not exist.

Bill



________________________________________
From: Joshua Wiley [jwiley.psych at gmail.com]
Sent: Thursday, October 07, 2010 9:14 PM
To: Schwab,Wilhelm K
Subject: Re: [R] Deleting observations - can't see the data after that

I mostly meant look at whatever past commands you had typed using the
"up" arrow (only useful for a very limited number).  I am sure this
has already taken enough of your time, and since its working for you
now, I would not worry about looking into it further.

I know that I have seen something in the past that produced a short
output about an object exactly like what you described (data frame
with n rows and m columns), but I cannot remember what it was for the
life of me.


On Thu, Oct 7, 2010 at 5:32 PM, Schwab,Wilhelm K <bschwab at anest.ufl.edu> wrote:
> Josh,
>
> Sounds good.  Where would I find the history?  I'm working on Linux (Ubuntu 9.10, R 2.9.2); if it's history(), we're out of luck.  You guys are allowed to hound; whether or not I can create a suitable example is another story.
>
> As far as what was happening, a summary of the object makes a lot of sense, and that's pretty much what it was.  Something like "a data frame with 16 rows and 5 columns" or there abouts.
>
> Bill
>
>
>
> -----Original Message-----
> From: Joshua Wiley [mailto:jwiley.psych at gmail.com]
> Sent: Thursday, October 07, 2010 7:22 PM
> To: Schwab,Wilhelm K
> Cc: r-help at r-project.org
> Subject: Re: [R] Deleting observations - can't see the data after that
>
> Dear Bill,
>
> We hound because we care---through repeated painful experiences, I have developed an avoidance to using function names for my functions/objects (and against irons near my fingers...but that is another story).
>
> If you still have the output from R when you attempted to print your data frame, I would be interested in seeing it.  It almost sounds like some sort of summary of the object, rather than the object itself (if that makes any sense).  Maybe its still in your history?
>
> As a side note, depending on the situation, you might get some mileage out of with() to lessen the this$that burden.  If you didn't know about it, hopefully it saves you at least a bit of time :)
>
> Here's to a better next two days than your last,
>
> Cheers,
>
> Josh
>
> On Thu, Oct 7, 2010 at 5:16 PM, Schwab,Wilhelm K <bschwab at anest.ufl.edu> wrote:
>> First, no lasting hard feelings - I've had two days of people riding me over minutia like you can't imagine.
>>
>> When you put this in the context of a possible bug, I'll see what I can turn up for you.  FWIW, I think it just the variable name.
>>
>> Bill
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: Peter Ehlers [mailto:ehlers at ucalgary.ca]
>> Sent: Thursday, October 07, 2010 7:10 PM
>> To: Schwab,Wilhelm K
>> Cc: r-help at r-project.org
>> Subject: Re: [R] Deleting observations - can't see the data after that
>>
>> On 2010-10-07 17:58, Schwab,Wilhelm K wrote:
>>> Foolish?  Try convenient.  Can't win for losing today.  Anyway, I most certainly did not make the mistake you suggest, though some other mistake is possible.  I never said it printed nothing; I was very explicit that it described it as a data frame with the correct number of rows and columns; it simply would not print the data.
>>
>> I didn't mean to be critical. I'm just trying to understand how you managed to get to the stage where R will show you that 'data' "is a data frame with specific (correct) number of rows and columns, but won't show me what remains in the frame".
>>
>> This should be reproducible. Who knows, you may have found a bug that should be fixed. So what was the precise message from R when it told you that it had the dataframe but wouldn't print it. Can you make up a reproducible example?
>>
>>   -Peter Ehlers
>>
>>>
>>>
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Peter Ehlers [mailto:ehlers at ucalgary.ca]
>>> Sent: Thursday, October 07, 2010 6:53 PM
>>> To: Schwab,Wilhelm K
>>> Cc: r-help at r-project.org
>>> Subject: Re: [R] Deleting observations - can't see the data after
>>> that
>>>
>>> On 2010-10-07 17:13, Schwab,Wilhelm K wrote:
>>>> Josh, Jim,
>>>>
>>>> Thanks for responding.  So far, it looks like my use of the name data was the problem - that could have taken some time to find.  I typically do not attach frames (and did not here), so I end up with lots of this$that in my code.
>>>>
>>>
>>> While I think it's foolish to call your data.frame 'data', I really
>>> doubt that that's the cause of your troubles. More likely you did
>>> something else afterwards that caused your data to be 'unprintable'.
>>> Or perhaps you goofed up the subsetting with something like
>>>
>>>    data = data(-3,);
>>>
>>> But I would have expected R to print _some_ thing, if only an error message.
>>>
>>> Anyway, I'm glad the problem is resolved (for now).
>>>
>>>     -Peter Ehlers
>>>
>>>
>>>> If it gives me any more trouble, I will indeed post an example.
>>>>
>>>> Thanks!
>>>>
>>>> Bill
>>>>
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Joshua Wiley [mailto:jwiley.psych at gmail.com]
>>>> Sent: Thursday, October 07, 2010 4:46 PM
>>>> To: Schwab,Wilhelm K
>>>> Cc: r-help at r-project.org
>>>> Subject: Re: [R] Deleting observations - can't see the data after
>>>> that
>>>>
>>>> Hi Bill,
>>>>
>>>> Several things come to mind.  First, try naming your data frame something besides a function name (data() is also a function).
>>>> Second, have you attached the data frame?
>>>>
>>>> Using: data = data[-3, ] worked fine for me when I made up some data.
>>>> Perhaps you can create a minimal and reproducible example?
>>>>
>>>> You might also send us the results of:
>>>>
>>>> sessionInfo()
>>>> ls()
>>>> search()
>>>>
>>>> Cheers,
>>>>
>>>> Josh
>>>>
>>>> On Thu, Oct 7, 2010 at 2:30 PM, Schwab,Wilhelm K<bschwab at anest.ufl.edu>   wrote:
>>>>> Hello all,
>>>>>
>>>>> I am loading a data frame, fitting a model, getting diagnostic plots and they are flagging a couple of observations as problematic.  Fair enough, and I want re-fit without them.
>>>>>
>>>>> After I delete an offending row (identified by one of the
>>>>> diagnostic plots), something like
>>>>>
>>>>>      data = data[-3,];
>>>>>
>>>>> then R will no longer print the contents of the data frame; it tells me it is a data frame with specific (correct) number of rows and columns, but won't show me what remains in the frame like it does before the deletion.  Is there a way to get around that, either using a different deletion technique or another function?  print(data) and show(data) are not helping.
>>>>>
>>>>> Ultimately,  I am trying to go through a couple of iterations of find pathologic points, delete and re-fit.  In this case I could guess at what is wrong and probably be correct, but I want to follow the clues as a learning exercise.  Once that is complete, I plan to plot everything with the deleted points emphasized.
>>>>>
>>>>> Bill
>>>>>
>>>>> ______________________________________________
>>>>> R-help at r-project.org mailing list
>>>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>>>> PLEASE do read the posting guide
>>>>> http://www.R-project.org/posting-guide.html
>>>>> and provide commented, minimal, self-contained, reproducible code.
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Joshua Wiley
>>>> Ph.D. Student, Health Psychology
>>>> University of California, Los Angeles http://www.joshuawiley.com/
>>>>
>>>> ______________________________________________
>>>> R-help at r-project.org mailing list
>>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>>> PLEASE do read the posting guide
>>>> http://www.R-project.org/posting-guide.html
>>>> and provide commented, minimal, self-contained, reproducible code.
>>>
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> --
> Joshua Wiley
> Ph.D. Student, Health Psychology
> University of California, Los Angeles
> http://www.joshuawiley.com/
>



--
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/



More information about the R-help mailing list